Dissecting and tracking socio-spatial
disadvantage in urban Australia
Hal Pawson, City Futures Research Centre, University of New South Wales
Special acknowledgements to:
 Shanaka Herath, University of New South Wales
 Kath Hulse and Margaret Reynolds, Swinburne University
 George Galster
Background/context
 Paper draws on analysis of spatially
concentrated disadvantage in Sydney,
Melbourne and Brisbane
 Place-based disadvantage recognised as
priority in social inclusion agenda
 Key underpinning research hypotheses:
a) Disadvantaged places have diverse social
and housing market structures, and
b) Appropriate policy responses vary
accordingly
 Builds on Australian ‘geography of poverty’
tradition going back to 1970s
 Presentation focuses on outputs from stats
analysis—framing structure for primary
fieldwork
1
Background/context
 Australia’s income distribution less
skewed than UK/US—but rising
inequality post-90s
 De-industrialisation limited compared
with many other OECD countries
 Little post-industrial heritage of
blighted localities or regions
 Ongoing migration-fuelled population
growth—approx 2% p.a.
 Highly urbanised—Sydney,
Melbourne and Brisbane = 50% of
national population
 96% of housing privately owned
2
Research questions and
presentation structure
1. What is the (people-based) geography of
disadvantage across the three cities?
2. How can we understand and capture heterogeneity
of disadvantaged areas?
3. How is the geography of disadvantage changing
over time and what are the policy implications?
Step 1
Pinpointing disadvantaged places
 Population-based concept of
disadvantage adopted
 Reference made to ABS censusbased index of deprivation—SocioEconomic Indicator for Areas
(SEIFA)
 Analysis based on CDs (avg popn:
600) and suburbs (avg popn: 6,000)
 ‘Disadvantaged area’ threshold:
lowest SEIFA quintile, nationally
 Variant analysis used SEIFA lowest
decile threshold
Key outputs of analysis
Sydney, Melbourne, Brisbane
177
Disadvantaged suburbs
10%
Population in disadv
suburbs as % of total
city population
16
% of disadv CDs in
disadvantaged suburbs
72
Disadvantaged suburbs
in Sydney, 2006
SEIFA quintile and decile
thresholds
 In all three cities, disadvantaged
suburbs clustered in middle and
outer urban locations
 In Sydney, agglomerations to
W, SW and far NE of metro area
 Red areas show lowest decile
disadv suburbs
 Geography of disadvantage
radically reshaped since 1980s
 Key role of inner area
gentrification
 Rental housing disproportionate,
but public housing only 13%;
home ownership: 54%
Disadvantaged suburbs
in Melbourne, 2006
SEIFA quintile and decile
thresholds
Disadvantaged suburbs
in Brisbane, 2006
SEIFA quintile and decile
thresholds
Typical low-income housing
forms in Sydney
Step 2
Classifying disadvantaged places
 Suburb-level census
data analysis of:

Residential mobility
 Demographic profile
 Area socio-economic
trajectory
 Cluster analysis
identifies common
permutations in
indicator scores
 Four distinct
disadvantaged area
categories emerged
Dimension
Indicator(s)
A
Residential
mobility
B
Lifecycle
stage/demo
-graphic
profile
% of hhlds moved in last 5 years
% of hhlds moved from overseas in last 5
years
% over 65s not in labour force
% young people
% single parent families
% couples with dependent children
% lone person households
C
Socioeconomic
status
change
over time
2001–11 % pt change - unemployment rate
2001–11 % pt change - early school
leavers
2001–11 % pt change - 15–24 NEETs
2001–11 % pt change - low income hhlds
2001–11 % pt change - low skilled workers
Typology results
Summary
Distinguishing socioeconomic characteristics
Housing market designation
High on young people & single
parent households
‘Isolate suburbs’
High social rental; median sales
prices and rents far below citywide norms
Type 2
High on overseas movers
‘Lower price suburbs’
Relatively affordable house prices
and distinct low rent market
Type 3
High on residential mobility,
(domestic movers), high on older
people
‘Marginal suburbs’
Remote from mainstream markets;
high concentration of low sales
prices & rents
Type 4
High on overseas movers, high
on reduced unemployment &
incidence of low status jobs
‘Dynamic improver suburbs’
Sales prices and rents moving
rapidly towards city-wide norms
Type 1
Disadvantaged suburbs
in Sydney, 2006
Socio-economic typology
 Disadvantaged area types
highly spatially grouped and/or
spatially distinctive
 Type 1 area distribution
influenced by public housing
geography
 Type 3 areas v peripheral
 Types 2 and 4 distinguished
by accessibility
 Observations also hold true
for patterns in Melb and
Brisbane
Disadvantaged suburbs
in Melbourne, 2006
Socio-economic typology
Disadvantaged suburbs
in Brisbane, 2006
SEIFA quintile and decile
thresholds
Change over time 2006–11
 SEIFA rankings 2006 and
2011 compared to analyse
change over time
 Socio-spatial segregation
continuing to intensify as
shown by:
Rising % of disadv small
areas in disadv suburbs
 Faster increase in ‘most
disadv’ suburbs
Measures of change, 2006–11
% of suburbs disadv 2006
10
% of suburbs disadv 2011
11
Absolute change 2006-2011
+1
% of disadv CDs in disadv suburbs 2006
72
% of disadv SA1s* in disadv suburbs 2011
74
Absolute change 2006–11
+2
% chg in no of disadv suburbs—quintile
threshold
+6
% chg in no of disadv suburbs—decline
threshold
+22

The spatial anatomy of
change over time
 Net change in no of disadvantaged suburbs reflects dynamic
process—not just minor addition to existing disadvantaged cohort
 ‘Becoming disadvantaged’ areas include disproportionate no of
suburbs akin to ‘isolate’ and ‘marginal’ areas—local housing
market conditions implicated as underlying driver of evolving
social geography
 Suburbs akin to Type 4 (‘dynamic improver’) areas
overrepresented within ‘ceasing to be disadvantaged’ cohort
Sydney, Melbourne and Brisbane
Avg distance from
CBD (km)
All disadv
suburbs 2006
All disadv
suburbs 2011
Ceased to be
disadv 2011
Became disadv
2011
28.7
30.2
23.1
35.6
International
contextualisation
 Post-2000 suburbanisation of
disadvantage a general trend in US cities:

poor population ↑53% 2000–10 in suburbs
of 100 largest US metro areas: cities ↑23%
(Berube & Kneebone 2011)
 Disproportionate rise in suburban poverty
in Canada post-1970—‘The Three Cities
of Toronto’ (Hulchanski 2011)
 UK—similar dynamics moderated by
more extensive social housing but (esp in
London) catalysed by post-2010 welfare
reforms
 Inner city (high value) social housing
disposal debated
Suburbanisation of
disadvantage: does it matter?
 Less mobile populations increasingly in less accessible places—
potentially remote from jobs and services
 General suburbanisation of employment in US means reduced
access to services possibly more problematic
 In Australia’s mono-centric cities CBDs and inner areas continue
to dominate ‘knowledge economy’ job growth [next slide]
 Sydney: jobs growth in ‘global arc’ 2.1% pa but disadv
population increasingly dispersed to Western Sydney—job
growth 0.5% pa Without assertive employment planning policy,
Australia’s major cities face policy choice:

Protect and promote inner area affordable housing (e.g. via assertive
inclusionary zoning) or
 Large-scale transport investment to facilitate greater commuting flows
Disadvantaged suburbs
in Sydney
Change over time 2006–11
Summary
 Distinct types of disadvantaged places can be
identified—likely to differ in extent to which ‘a policy
problem’
 Strong connections between local housing market
conditions and area socio-economic profiles/trajectories
 Disadvantage becoming more polarised and more
suburbanised
 Especially problematic in mono-centric cities where that
very mono-centrism is a driving force of the process itself
References
Berube, A. & Kneebone, E. (2011) Parsing U.S. Poverty at the Metropolitan Level;
Washington: Brookings Institution
Hulse, K. Pawson, H. Reynolds, M. & Herath, S. (forthcoming 2014) Disadvantaged places
in urban Australia: analysing socio-economic diversity and housing market performance;
Melbourne: AHURI
Hulchanski, D 2010, The three cities within Toronto: income polarization among Toronto’s
neighbourhoods, 1970–2005; Toronto: University of Toronto
Pawson, H. Davison, G. & Wiesel, I. (2012) Addressing concentrations of disadvantage:
policy, practice and literature review, Final Report No. 190, Australian Housing and
Urban Research Institute, Melbourne
Pendall, R., Weir, M. & Narducci, C. (2014) Governance and the Geography of Poverty:
Why Does Suburbanization Matter? Working paper; MacArthur Foundation Network,
University of California, Berkeley
Randolph, B. & Holloway, D. (2005), ‘The suburbanization of disadvantage in Sydney: new
problems, new policies’, Opolis, 1(1) pp.49-65
Randolph, B. & Tice, A. (2014) Suburbanizing disadvantage in Australian cities: sociospatial
change in an era of neoliberalism; Journal of Urban Affairs, 36 (S1) pp1–16